They trained the model (U-Net) with frames from numerical solutions of the hydrodynamical equations.
Today, we are bringing other exciting results involving black holes and AI. We released a new paper:
"Black Hole Weather Forecasting with Deep Learning: A Pilot Study"
Work by Roberta Duarte (@import_robs), Rodrigo Nemmen (@nemmen) and João Paulo Navarro (from @NVIDIABrasil).
They trained the model (U-Net) with frames from numerical solutions of the hydrodynamical equations.
We want to investigate if deep learning can be a new method to simulate accurately in less time!
1- The model simulating only one system after learning only from this system
2- The model simulating an unseen system after training with several systems with different initial conditions
The result is that the model can simulate up to 8e4 gravitational time accurately with a speed-up of 30000x faster!
However, they hid one system to understand the generalization power of the model!
It simulated the unseen system for 4e4 gravitational time, showing that the model can generalize the black hole physics presented in the dataset!
More from Science
I took a look at Shell's first ever 1.5C scenario and found that it is... remarkably similar to its “well-below 2C” scenario.
Oil, gas, coal, solar.... all basically unchanged.
The key difference: A new forest the size of Brazil to suck up the extra CO2.
Including "nature-based solutions" in the outlook brings forward the date for net-zero emissions to 2058.
Without them their pathway for CO2 emissions is the same as the previous one.
(It's also towards the higher end of 1.5C emissions pathways.)
The "Brazil-sized" forest idea isn't actually new, it has been kicking around for a couple of years.
It was referenced in the "well-below 2C" scenario although not formally included in it, and Shell's CEO has been framing it as the only viable way of getting to 1.5C.
Fine, but who is going to plant all those trees? Well... Shell says it will plant some of them.
Only yesterday Shell said forests were a key part of its net-zero strategy.
Not everyone is convinced though
https://t.co/RaJm7tOHxb
Given that Shell's 1.5C scenario also sees a big scaling up of bioenergy, the question remains: where are all those trees and bioenergy crops going to go?
Oil, gas, coal, solar.... all basically unchanged.
The key difference: A new forest the size of Brazil to suck up the extra CO2.
Including "nature-based solutions" in the outlook brings forward the date for net-zero emissions to 2058.
Without them their pathway for CO2 emissions is the same as the previous one.
(It's also towards the higher end of 1.5C emissions pathways.)
The "Brazil-sized" forest idea isn't actually new, it has been kicking around for a couple of years.
It was referenced in the "well-below 2C" scenario although not formally included in it, and Shell's CEO has been framing it as the only viable way of getting to 1.5C.
Fine, but who is going to plant all those trees? Well... Shell says it will plant some of them.
Only yesterday Shell said forests were a key part of its net-zero strategy.
Not everyone is convinced though
https://t.co/RaJm7tOHxb
Shell plans to use forests to remove 120 Mt/yr of CO2 by 2030.
— Greg Muttitt (@FuelOnTheFire) February 12, 2021
Appropriate land for forestation is finite, and risks competition with food production and human rights of current land owners/users, esp Indigenous
Given that Shell's 1.5C scenario also sees a big scaling up of bioenergy, the question remains: where are all those trees and bioenergy crops going to go?
